An integrated novel triangulation framework for monitoring ecosystem health and sustainability under regenerative landscapes

Abstract

Accelerated climate change and unsustainable agricultural practices have intensified greenhouse gas (GHG) emissions and degraded soil health, particularly in intensively cultivated landscapes. This study presents an integrated geospatial and biophysical triangulation framework to monitor ecosystem health and sustainability in the rice-based systems of Chhindwara district, Madhya Pradesh, India. The objectives were to evaluate GHG emissions, assess land use change, analyze soil health, and identify socio-economic drivers influencing agricultural sustainability under a regenerative agriculture (RA) paradigm. A combination of high-resolution satellite imagery, field-based soil sampling (n = 430), socio-economic surveys, and carbon stock modeling using the InVEST tool was employed. Land Use and Land Cover (LULC) changes over a decade (2011–2021 years) were mapped using Random Forest classification. GHG emissions were estimated using the Cool Farm Tool, while Water Use Efficiency (WUE) and biodiversity indices were evaluated across villages and seasons. In Sausar, there was a 4.6% decrease in agricultural area, a 6.6% rise in dense vegetation, and noticeable soil degradation. Mokhed had a higher soil organic content (1.07%) than Sausar (0.51%), and its carbon stocks were higher (13–35 Mg C/ha) than Sausar’s (7–13 Mg C/ha). Furthermore, Mokhed’s GHG emissions during the Kharif season were much greater at 4,000 kg CO₂ eq/ha than those during the Rabi season, which were just 1,500 kg CO₂ eq/ha. WUE varied by season and crop, with Rabi vegetables achieving the maximum WUE at 10.4 kg/ha-mm and cotton demonstrating the lowest efficiency at 1.4 kg/ha-mm in Kharif, underscoring the potential of targeted interventions. The framework demonstrated that integrating geospatial, biophysical, and socio-economic data provides actionable insights for climate-resilient agriculture. The findings support landscape-level planning for soil restoration, GHG emission mitigation, and sustainable intensification in rainfed regions to escalate RA transitioning and benefits for the local communities.

1 Introduction

Agroecosystems play a crucial role in striking a balance between meeting human needs and preserving the harmony of nature (Marcacci G. et al., 2025). However, the sustainability of agro-ecological landscapes faces threats from climate change, food insecurity, and a range of systemic issues, including industrial farming, chemical dependency, environmental degradation, and corporate dominance (Padhiary and Kumar, 2024; Willett et al., 2019). These challenges, compounded by greenhouse gas emissions and soil degradation, undermine livelihood security. A holistic shift from conventional agriculture (CA) to regenerative agriculture (RA) is essential to address these issues and harness the benefits of landscapes (Campbell et al., 2017). In the agenda for Sustainable Development Goals (SDG) 2030, food and agriculture are crucial, aimed at ending poverty and hunger (SDG1 and SDG2), addressing climate change (SDG13), and sustaining natural resources. Though, failure to reduce hunger and malnutrition hinders progress on SDGs (Poore and Nemecek, 2018). The growing population, climate change, water shortages, and land degradation highlight the need to boost agricultural production, especially in rainfed areas (Godfray et al., 2010). About 80% of the required increase must come from intensifying existing systems to meet targets without extensive land conversion (Campbell et al., 2017). The agriculture sector faces the challenge of feeding a growing population with limited water, land, and biodiversity resources. Currently, the global food system contributes 25% of annual greenhouse gas emissions, one-third of terrestrial acidification, and most surface water eutrophication (Poore and Nemecek, 2018). Continuing unsustainable practices, such as synthetic pesticide use, artificial fertilizers, fossil fuels, and food waste, risks exceeding the planet’s carrying capacity (Willett et al., 2019). Although producing food within the planet’s carrying capacity is a growing priority, reflected in policies like EU Circular Economy Action Plan (European Commission, 2015), the Paris Climate Agreement (United Nations, 2015), and the Common Agricultural Policy (European Commission, 2019). Faulty land management has caused 50%–70% of agricultural soil organic carbon (SOC) lost (Lal, 2010), significantly reducing the soil’s carbon storage capacity of 248 Pg in the top 3 m (Koglo et al., 2016). Improved agronomic practices, precision input management, and efficient resource use can help to mitigate greenhouse gas emissions. In this context adopting Regenerative agriculture (RA) is crucial not only for improving human health and economic prosperity but also for conserving cultural ecosystem services. Though, many researchers conducted individual studies on soil health and GHG emissions (Battaglia et al., 2021; Yang et al., 2024; Sroka et al., 2019). Geospatial technologies are increasingly crucial for understanding landscape-level transformations in agroecosystems, particularly under the regenerative agriculture (RA) paradigm. High-resolution satellite remote sensing, when combined with ground-based observations and geostatistical modeling, enables the continuous, cost-effective monitoring of spatial and temporal changes in land use, vegetation dynamics, and ecological health (Chaudhuri and Mishra, 2016; Pandey et al., 2018). In this context, satellite-based spatial and temporal variability of land use/land cover (LULC) assessments provide robust insights into land degradation, forest transitions, and cropland dynamics over time, which are critical for evaluating the ecological impacts of regenerative interventions. Moreover, geospatially enabled tools such as the InVEST model allow for spatial quantification of ecosystem services including soil organic carbon (SOC) storage and carbon sequestration across diverse agro-ecological contexts (Richard et al., 2020).

Despite the growing use of these tools, integrated geospatial frameworks that connect LULC trajectories with biophysical (e.g., soil, water, carbon) and socio-economic indicators remain scarce in RA literature (Mishra et al., 2024). This study bridges that gap by deploying a triangulated geospatial framework combining multi-temporal satellite imagery, machine learning-based classification, and carbon modeling to assess landscape health and sustainability transitions (Potapov et al., 2022; Zou et al., 2022). By integrating these outputs with soil carbon simulations from InVEST and Green House Gas (GHG) estimation via FIRMS (Fire Information for Resource Management System)-based fire detection and the Cool Farm Tool, this framework delivers spatially explicit evidence of regenerative transitions, making it highly relevant for MRV (Monitoring, Reporting and Verification) systems, carbon financing, and sustainable land-use policy (Hillier et al., 2011; Badarinath et al., 2009).

However, no integrated study was conducted on the Geospatial, bio-physical, and socio-economical triangulation for monitoring ecosystem health and sustainability with special reference to the cotton production associated with the RA Framework (Figure 1). Hence, in this research an integrated novel approach for the RA baseline framework was undertaken (i) to analyze spatial and temporal variability in land use and land cover, highlighting changes over time. (ii) to characterize the region based on socio-economic factors, agricultural and livestock practices, and climate change impacts on farming. (iii) to evaluate the impact of current agricultural practices on environmental sustainability, focusing on agronomic efficiency, carbon sequestration, and soil degradation across landscapes, and (iv) to estimate baseline greenhouse gas emissions from various landscapes to inform mitigation strategies. This holistic approach offers a baseline for assessing RA, helping researchers identify diversification and soil management options to improve soil health and sustainability.

FIGURE 1

Agroecosystems are central to sustaining food security, ecosystem services, and rural livelihoods, particularly in regions grappling with socio-economic fragility and climatic uncertainty. However, contemporary agricultural systems have been increasingly characterized by unsustainable intensification marked by excessive agrochemical usage, over-reliance on monocultures, and poor land stewardship resulting in degraded soils, declining water quality, biodiversity loss, and heightened greenhouse gas (GHG) emissions (Lal, 2010; Poore and Nemecek, 2018). These trends have profound implications for long-term agroecological resilience, particularly in the Global South where smallholder agriculture dominates. In India, for instance, the degradation of soil organic carbon (SOC), unbalanced nutrient cycles, and water inefficiencies have collectively impaired the regenerative capacity of agricultural landscapes (Padhiary and Kumar, 2024).

The need for a transformative shift in agriculture is increasingly being recognized, leading to the emergence of regenerative agriculture (RA) as a comprehensive and holistic approach to address the interlinked crises of food insecurity, climate change, and environmental degradation. Unlike conservation or climate-smart agriculture, regenerative agriculture emphasizes not only sustainability but also the restoration and enhancement of ecosystem functionparticularly soil health, biodiversity, and carbon sequestration through nature-based and low-external-input practices (Newton et al., 2020; Dabalen et al., 2024). Its core principles include minimizing soil disturbance, maximizing biodiversity, maintaining continuous soil cover, integrating livestock, and recycling organic nutrients (Khangura et al., 2023; Giller et al., 2021).

One of the distinguishing features of RA lies in its system-thinking approach that integrates soil-plant-microbe interactions with above- and below-ground biodiversity and socio-economic resilience. By reconfiguring land-use practices from extractive to regenerative, RA holds the potential to simultaneously enhance yields, reduce input costs, and improve ecological integrity generating what some scholars refer to as “triple wins”: productivity, resilience, and climate mitigation (Dabalen et al., 2024).

Despite this potential, there is still a paucity of integrated evidence at landscape scales. Most existing research has been fragmented, emphasizing either biophysical benefits (e.g., SOC sequestration, GHG reduction) or localized socio-economic improvements (e.g., income stability). This research seeks to address this critical gap by operationalizing a multi-dimensional triangulation framework for monitoring regenerative agriculture at landscape scales. We present empirical findings from Chhindwara district in Madhya Pradesh, India a region emblematic of rainfed agroecologies with mixed cropping systems including cotton and rice. The study leverages high-resolution satellite imagery, laboratory-based soil diagnostics, village-level livelihood surveys, and spatial modeling to develop an integrated baseline for evaluating RA impacts.

The rationale for selecting this landscape stems from its ecological vulnerability high SOC loss, erratic rainfall, land degradation and socio-economic diversity. These characteristics mirror the broader challenges faced across central India and many other developing regions where smallholder farmers are caught between ecological collapse and economic insecurity (Giller et al., 2021; Willett et al., 2019). Central to this analysis is the recognition that RA is not a one-size-fits-all model. Its outcomes are context-specific and influenced by local agroecological conditions, governance frameworks, and institutional capacity.

Another dimension of this study pertains to the evaluation of GHG emissions under alternative land management regimes. Drawing on tools like the Cool Farm Tool and InVEST SOC models, this work estimates seasonal emissions and mitigation potential from RA practices such as no-tillage, diversified cropping, and organic inputs. These insights are crucial in quantifying the climate benefits of regenerative transitions and contributing to MRV (Monitoring, Reporting, and Verification) frameworks under climate finance mechanisms. A key insight from recent empirical reviews is that regenerative outcomes are not only contingent on biophysical variables but also mediated by socio-economic structures access to markets, knowledge systems, gender roles, and local governance (Dabalen et al., 2024; Newton et al., 2020). For example, while women farmers often manage composting, seed saving, and water harvesting, their participation in decision-making remains limited due to entrenched gender norms. Recognizing and addressing these social dimensions is essential to ensure equitable scaling of RA.

2 Materials and methods

2.1 About study area and soil sampling

For this study, two blocks, Sausar and Mokhed, in the Chhindwara district of Madhya Pradesh, India, were selected, encompassing an area of 1,833 square kilometers (Figure 2). These blocks represent diverse agro-ecological zones with significant agriculture. Mokhed is primarily known for wheat, maize, vegetables, soybean, and horticulture, while Sausar is primarily known for cotton and orange while other crops are corn. Both blocks practice mixed farming, integrating crops and livestock. The Soil properties in the Mokhed and Sausa block are shown in Table 1.

FIGURE 2

All the soil samples were collected from 400 agricultural sites and 30 forest locations at various surface (0–15 cm) and subsurface (15–30 cm) depths. Nearly 30% area of Mokhed and Sausar are under forest coverage while in Mohkhed nearly 70% area are cultivable while in Sausar only 58.55 percent area is arable for cultivation. Stratified soil sampling locations is shown in Figure 3. Each geotagged sample was labeled with the village name, farmer’s name, depth, and GPS waypoint ID to ensure accurate tracking and analysis.

FIGURE 3

2.2 Geospatial assessment

2.2.1 Ground truth and satellite data collection

Approximately 1,150 ground truth points were collected to train and validate the satellite images for generating the Land Use Land Cover (LULC) map. Geo coordinates of land and feature-specific information were gathered using GPS (Garmin Oregon 650) and KOBO administered via questionnaires, respectively (Figure 4).

FIGURE 4

The study area includes two tiles of Resourcesat-1 and 2 satellite data (Figure 5), using cloud-free LISS-IV images with high spatial resolution (5.8 m) from 2011–12, 2016–17, and 2021–22. The images, acquired from NRSC Hyderabad, India, ensure temporal consistency and minimal cloud cover for accurate analysis. The selected years allow analysis of land cover changes and long-term trends over a decade in the study area.

FIGURE 5

2.2.2 Characterizing and mapping the LULC

The flowchart for the LULC classification methodology is shown in Figure 6. High-resolution imagery provides details land cover data, improving classification accuracy. A total number of 29 cloud-free images across 3 years and cropping seasons were used for analysis. Principal component analysis (PCA) were applied to reduce the redundancy in spectral bands, extracting key information from stacked images. In addition, the Random Forest classifier was implemented on the PCA-transformed images using the “RandomForest” package in R. The ground truth data for the year 2020–21 used as training datasets, while, multi-temporal Google Earth imagery was used to select suitable training sites. Data was split 70% for training and 30% for testing for robust validation.

FIGURE 6

The optimized Random Forest classifier was deployed to classify the study area for the years 2020–21, 2016–17, and 2011–12. The model performed three classification levels: first, separating vegetation, non-vegetation, and agricultural lands, and second, identifying nine LULC classes such as cropland, sparse vegetation, dense vegetation, water bodies, shrubland, wasteland, grassland, settlements, and sand/mining areas.

2.2.3 Carbon stock modelling

The Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model was utilized to assess soil carbon stock in the study area. The application of the InVEST model in this study enabled a robust, data-driven approach to evaluating soil carbon dynamics across diverse landscapes. The InVEST model provides a spatially explicit, data-driven framework for quantifying and mapping carbon storage and sequestration across heterogeneous landscapes. Specifically, the InVEST Carbon Storage and Sequestration module was employed, which integrates geospatial inputs such as land use and land cover (LULC) data, soil properties, and carbon pool estimates (InVEST®: https://naturalcapitalproject.stanford.edu/software/invest, Richard et al., 2020).

2.3 Socio economic survey

A detailed bilingual questionnaire was designed for data collection, including farmer interviews, and soil sampling. One-on-one interviews, and focus group discussions for accuracy, were also conducted to validate village data.

2.4 Bio-physical indicators

Bio-physical indicators are essential in regenerative agriculture, providing critical metrics to evaluate the health and sustainability of agricultural systems. These indicators include key parameters such as soil health, water use efficiency, and biodiversity, which collectively reflect the ecological integrity and productivity of farming practices.

2.4.1 Soil analysis

The collected soil samples were dried and sieved through a 2-mm mesh. The samples were subject to analysis for chemical (organic carbon content, macronutrient and key micronutrient), physical (bulk density) and biological (soil microbial biomass carbon and soil respiration) properties. Using standard methods and protocols, soil texture was analyzed using the hydrometer method (Gavlak et al., 2005), while soil pH and EC were determined from the supernatant liquid (Jackson, 1973). Soil organic carbon (SOC) was measured following Nelson and Sommers (1996), and bulk density was assessed according to Blake et al. (1986). Available nitrogen, phosphorus, and potassium were analyzed using the alkaline KMnO4 method (Subbiah and Asija, 1956), Olsen’s 0.5 M NaHCO3 extractable method (Olsen et al., 1954), and ammonium acetate exchangeable potassium method (Toth and Prince, 1949), respectively. Soil respiration was evaluated using the rapid titration method (Angers and Recous, 1997). All analyses were conducted at the Soil Laboratory of IRRI-South Asia Regional Center in Varanasi, India.

2.4.2 GHG estimation

The assessment of greenhouse gas (GHG) emissions was conducted using the open-source software Cool Farm Tool, a methodology aligned with the approach described by Hillier et al. (2011). This tool incorporates a combination of global empirical models, utilizing Tier 1, Tier 2, and simplified Tier 3 approaches to estimate farm-gate emissions both per unit area and per unit of product. Additionally, crop burning events were identified as a potential source of GHG emissions. To quantify these incidents, geospatial technology was employed to record the number of fire occurrences in agricultural lands. Crop residue burning is a significant source of greenhouse gas (GHG) emissions, contributing to air pollution and climate change, so using geospatial technology the fire incidents are monitored and quantified in agricultural lands, providing critical data for understanding the spatial and temporal patterns of crop burning (Badarinath et al., 2009).

2.4.3 Water use efficiency (WUE)

WUE is the biomass or grain produced per unit of water the crop uses. It is vital for sustainable agriculture and was calculated as the crop yield ratio to water use per unit area (Equation 1) for all key crops at the project site.

Where crop yield, in kilograms per hectare (kg/ha), and water input, in millimeters (mm) of water applied.

2.4.4 Biodiversity assessment at landscape

Biodiversity indicators at the landscape level were assessed through a structured socio-economic survey conducted across representative farming households in the study area. A stratified sampling approach was employed to capture variability across agro-ecological zones, with specific focus on the adoption of diversified agricultural practices contributing to on-farm biodiversity. Data on five key biodiversity-linked components agroforestry, livestock integration, vermicomposting, fisheries, and beekeeping were recorded using standardized questionnaires and validated through field verification. The frequency of practice adoption was quantified and expressed as a percentage of total respondents, enabling landscape-level benchmarking. The resulting statistics reflect the extent of integration of biodiversity-enhancing practices within the prevailing farming systems and serve as proxies for ecological sustainability, resilience, and multifunctionality of agroecosystems.

2.5 Analytical methods and statistical analysis

The data analysis was conducted using STATA v15, starting with descriptive statistics to summarize data on soil samples, GHG emissions, and SOC measurements. Linear and logistic regression models were used to examine relationships between variables, predicting SOC levels and crop yield based on soil properties and management practices. Application of analysis of variance (ANOVA) compared the soil health and GHG emissions across groups, while Kruskal–Wallis tested non-parametric data. Time series analysis identified trends and seasonal patterns in SOC and WUE over multiple seasons. Moreover, Principal Component Analysis (PCA) was applied to reduce the data dimensionality and highlight key variables. For the data visualization scatter plots, box plots, and histograms were used, and all analyses were documented in a STATA DO file for transparency and reproducibility.

A detailed assessment of the ecological and socio-economic implications of regenerative agriculture served as the basis for approach and parameter selection. The selection of each measure was based on how well it aligned with the main goals of RA, which are increased biodiversity, carbon sequestration, water use efficiency, and soil health. A thorough, multifaceted evaluation of RA’s efficacy at the landscape scale is provided by the study’s integration of sophisticated GIS techniques, soil analysis, GHG estimation, and socioeconomic surveys as shown in Figure 7.

FIGURE 7

3 Results and discussion

3.1 Decadal LULC mapping and change detection

The LULC and change detection in the study area of various multi-temporal classes (cropland, sparse vegetation, dense vegetation, waterbody, shrubland, wasteland, grassland, settlement, and Sand/Mining) for the years 2011–12, 2016–17, and 2020–21 are shown in Figures 8a–c. High-resolution LULC mapping using the Random Forest method captured spatiotemporal dynamics over 2011–2021. The maps indicated a significant 6.6% increase in dense vegetation, with transitions from shrublands to dense and sparse vegetation (Figure 8d). The cropland observed to decline by 4.6% over the past decade. Moreover, wasteland and shrublands declined sharply by 3.3% and 5.8%, respectively, indicating expansion in tree cover, grasslands, sand/mining, and urban areas. A similar pattern of spatio-temporal changes in vegetation was also observed by Zou et al. (2022) and Potapov et al. (2022). Graphical representation of the LULC class’s percentage change in areas over 10 years (2011–2021) is shown in Figure 9. The application of Random Forest algorithm achieved high overall accuracy for LULC maps: 91% (2020–21), 87% (2016–17), and 85.6% (2011–12).

FIGURE 8

FIGURE 9

3.2 Impact of biophysical indicators

The research data in Table 2 indicate that the average bulk density of soil in Mohkhed was 1.40 g/cm³ at a depth of 0–15 cm and 1.51 g/cm³ at 15–30 cm. Similarly, in Sausar, the bulk density values were 1.41 g/cm³ at 0–15 cm and 1.52 g/cm³ at 15–30 cm. These values exceed the ideal range, potentially hindering root growth. Whereas the soil pH of the study area was observed to be higher than 7 at both depths, indicating an alkaline nature. A soil pH between 6 and 7 is considered most favorable for nutrient absorption by plants (Marcacci M. et al., 2025). The value of EC was monitored in the normal range (<1 day/m), however, Sausar showed a higher EC value (0.5 ds/m) than Mokhed. The data indicated that Mokhed soils have higher organic content (1.1%) than Sausar (0.5%), with greater variation in soil organic carbon across villages of this block. The distribution of available soil nitrogen, pH and EC in both the block is shown in Figures 8a–c, indicated that high variability of nitrogen, with lower nitrogen levels in the southeastern Sausar block and higher levels in the northwest of Mokhed.

The soil carbon stock map was generated using ground carbon stock data, 2021 LULC data from LISS-IV, and the InVEST model. The distribution of carbon stock indicated the spatial variability of carbon in soil. Figure 9d illustrated that forest areas have higher carbon content, while agricultural areas show variable carbon levels. The mokhed block in the north has slightly higher carbon, compared to the lower carbon stock in the southern part of the study. The observation of the present study is good in line with the findings of Koga et al. (2020a), Koga et al. (2020b) and Jordon et al. (2022).

Cultivated land has half the SOC of forest soil, indicating soil degradation from intensive farming. SOC stock and available nitrogen in Sausar and Mokhed block is shown on Figure 10. The SOC stocks (0–30 cm) varied among the villages, ranging from 13 to 35 Mg C/ha (Mokhed) and 7–13 Mg C/ha (Sausar). This variation is due to differences in organic matter input, topography, vegetation, soil bulk density, moisture, and depth (Pathak and Reddy, 2021a, Pathak and Reddy, 2021b). By adding organic matter to the soil, microbes receive the carbon and nutrients they need, which immediately increases their activity and boosts the biomass and diversity of microorganisms overall. This increased biological activity enhances soil fertility, structure, and plant growth by improving processes such as nutrient cycling, stable soil aggregate formation, and improved water retention (Bot and Benites, 2005). Comparatively, the soil in Sausar block is more degraded than that in Mokhed and requires more intensive interventions to reverse the degradation process.

FIGURE 10

3.2.1 Estimation of GHG emissions

The analysis of greenhouse gas (GHG) emissions across two agricultural seasons, Kharif and Rabi, in the regions of Mohkhed and Sausar (Figure 11) reveals significant seasonal and locational variations. The total CO₂ emissions are consistently higher during the Kharif season compared to the Rabi season for both locations. Specifically, during the Kharif season, Mohkhed records emissions of approximately 4,000 kg CO₂ eq/ha, while Sausar exhibits slightly higher emissions at around 4,500 kg CO₂ eq/ha. In contrast, the Rabi season shows a marked reduction in emissions, with both locations reporting values near 1,500 kg CO₂ eq/ha. The observed seasonal inequality in emissions can be attributed to differences in agricultural practices, crop types, and environmental conditions prevalent during these periods. The Kharif season, typically associated with monsoon crops, may involve practices that contribute to higher GHG emissions, such as increased use of fertilizers, waterlogging, specific crop types with higher carbon footprints and the presence or absence of N₂O abatement technologies (Brentrup et al., 2004). Conversely, the Rabi season, characterized by winter crops, likely involves less intensive practices, resulting in lower emissions.

FIGURE 11

The box plot diagrams (Figures 12a–d) illustrate the total GHG emissions for each village in Mohkhed and Sausar. In Mohkhed, Patniya recorded the highest emissions during the Kharif season, averaging 6.1 mt CO₂/ha (Figure 12a) due to rice cultivation, while Sohagpur exhibited the highest emissions in the Rabi season, with an average of 5.1 mt CO₂/ha (Figure 12b). Conversely, Simariya Kalan and Chourai had relatively lower emissions during the Kharif and Rabi seasons, respectively. In Sausar, Amla showed the highest emissions in the Kharif season (5.3 mt CO₂/ha), owing to rice cultivation, whereas Koporwadi had the highest emissions in the Rabi season (2.8 mt CO₂/ha) (Figures 12c,d). Similar results of higher GHG emission during rice growing season were observed by Kabato et al. (2025). Kuddam village recorded the lowest CO₂ emissions across both seasons.

FIGURE 12

3.2.1.1 Burning incidents maps

The burning incident maps analyzed the fire incidents using data from NASA’s Fire Information for Resource Management System (FIRMS), specifically the SUOMI VIIRS C2 dataset spanning 2012 to 2021. Temporal trends in fire incidents were assessed by plotting the cumulative frequency of events for three representative years 2012, 2016, and 2021 focusing on the post-harvest periods of the Kharif and Rabi cropping seasons (Figure 13). Spatial analysis revealed a negligible number of fire incidents following the Kharif harvest, with a slight increase observed over the study period. Conversely, fire incidents following the Rabi harvest were significantly more frequent and exhibited a consistent upward trend across the years. The higher incidence of post-Rabi fires is likely attributable to farmers’ practices aimed at expediting field clearance for short-duration crops, leveraging the residual soil moisture available during this period.

FIGURE 13

3.2.2 Water use efficiency across different crops and seasons

Table 3 shows the analysis of WUE across Kharif and Rabi seasons revealed marked spatial and temporal differences driven by water availability, crop type, and management practices. In Kharif, rainfed systems such as cotton exhibited the lowest WUE (1.4 kg/ha-mm), while maize and rice showed moderate values (2.5–3.6 kg/ha-mm). In contrast, Rabi vegetables under irrigated regimes recorded the highest WUE (10.4 kg/ha-mm), underscoring the role of controlled water management in maximizing productivity (Passioura, 2006). Villages such as Rahap (Kharif) and Saroth (Rabi) in Mokhed, and Ghogharikhapa in Sausar, emerged as WUE benchmarks. The intra- and inter-block variability ranging from <1 to >12 kg/ha-mm reflects differing irrigation infrastructure, soil-moisture retention, and agronomic intensity. These findings reinforce the need for WUE optimization through micro-irrigation, mulching, and crop selection, particularly in low-performing zones. Targeted dissemination of WUE benchmarks and capacity-building interventions can bridge performance gaps and enhance water sustainability (Bossio and Geheb, 2010).

The crop-wise water productivity (kg/ha-mm) for different agricultural crops under varying seasonal conditions were presented in Figure 14. Cotton, primarily a rainfed Kharif crop, exhibits the lowest water productivity at 1.4 kg/ha-mm, with an effective value of 1.19 kg/ha-mm and maize, shows higher productivity at 3.6 kg/ha-mm (effective: 2.5 kg/ha-mm). While Rabi crop vegetables demonstrate the highest water productivity at 10.4 kg/ha-mm (effective: 7.1 kg/ha-mm). These variations highlight the influence of seasonal conditions and irrigation access on crop water productivity.

FIGURE 14

3.2.3 Biodiversity status at the landscape level

The landscape-level biodiversity audit revealed asymmetrical adoption of sustainable practices. Agroforestry (10%) and livestock integration (8%) were the most prevalent, reflecting partial alignment with ecological intensification principles (Figure 15). However, near-zero adoption of vermicomposting (2%), fisheries (0.2%), and beekeeping (0.0%) indicates critical gaps in pollinator support, nutrient cycling, and agroecological diversification (Jose, 2009; Garibaldi et al., 2013). The absence of apiculture undermines pollination-dependent crop productivity and landscape resilience. Moreover, limited vermicomposting uptake may reflect barriers in knowledge transfer and market connectivity. To transition toward functionally diverse, resilient agroecosystems, biodiversity-enhancing interventions must be mainstreamed through integrated extension models, incentives for ecosystem services, and landscape-scale ecological planning (Mijatović et al., 2013).

FIGURE 15

3.2.4 Limitations of the study

Some limitations of this work include the possibility of inaccuracies in remote sensing data from 29 cloud-free photos (2011–12, 2016–17, and 2020–21) due to cloud cover and spatial resolution (5.8 m), which could have affected the accuracy of Land use/land cover classification. The heterogeneity between remote areas may not be fully captured by soil sampling, even though it was conducted at 400 agricultural sites and 30 forest locations. With carbon stocks ranging from 7 to 35 Mg C/ha, the accuracy of the InVEST model’s carbon stock assessment depends on the correctness of input data, including land use and soil characteristics. The socio-economic survey, which relied on village-level interviews, might have been biased by respondents, and the study’s decade-long scope may not have fully captured current or longer-term changes in environmental and land use variables.

4 Conclusion

This study investigated biodiversity, water use efficiency (WUE), soil health, GHG emissions, and LULC variations in the Mohkhed and Sausar blocks of Chhindwara District from 2011 to 2021. Significant soil degradation was observed especially in Sausar, and there was a 6.6% increase in dense vegetation and a 4.6% decline in agricultural land. Mokhed had a higher soil organic content (1.07%) than Sausar (0.51%), and its carbon stocks (13–35 Mg C/ha) were higher than Sausar’s (7–13 Mg C/ha). In comparison to the Rabi season (1,500 kg CO₂ eq/ha), GHG emissions were higher during the Kharif season (4,000 kg CO₂ eq/ha in Mohkhed). Rabi vegetables demonstrated the best water use efficiency at 10.4 kg/ha-mm, while WUE varied among crops. Furthermore, it was found that biodiversity-enhancing activities like vermicomposting (2%), and beekeeping (0%), were not widely adopted. Policies should encourage farmers to implement biodiversity-enhancing methods, effective water management systems, and soil health restoration approaches in order to promote the adoption of regenerative agriculture (RA) practices. To encourage carbon sequestration activities and promote sustainable land management, governments should consider implementing carbon credit schemes. Long-term ecological and economic resilience can also be ensured by providing smallholder farmers with focused training and financial assistance to help scale RA techniques, especially in regions like Sausar where soil erosion is severe, thereby offering a replicable, science-based model for operationalizing RA and landscape restoration in smallholder agroecosystems under increasing climate stress.

References

• 1

  AngersD. A.RecousS. (1997). Decomposition of wheat straw and rye residues as affected by particle size. Plant and Soil, 189 (2), 197–203. 10.1023/A:1004270417878

  • CrossRef
  • Google Scholar

• 2

  BadarinathK. V. S.ChandT. R. K.PrasadV. K. (2009). Agriculture crop residue burning in the indo-Gangetic Plains—A study using IRS-P6 AWiFS satellite data. Curr. Sci.91 (8), 1085–1089.

  • Google Scholar

• 3

  BattagliaM.ThomasonW.FikeJ. H.EvanyloG. K.von CosselM.BaburE.et al (2021). The broad impacts of corn stover and wheat straw removal for biofuel production on crop productivity, soil health and greenhouse gas emissions: a review. Gcb Bioenergy13 (1), 45–57. 10.1111/gcbb.12774

  • CrossRef
  • Google Scholar

• 4

  BlakeG. R.HartgeK. H. (1986). Bulk density. In “Methods of soil analysis. Part 1: Physical and mineralogical methods“, A. Klute (Editor), 2nd ed. 363–375 (American Society of Agronomy and Soil Science Society of America).

  • Google Scholar

• 5

  BossioD.GehebK. (2010). Conserving land, protecting water. International Water Management Institute. Available online at: https://www.iwmi.cgiar.org.

  • Google Scholar

• 6

  BotA.BenitesJ. (2005). The importance of soil organic matter: key to drought-resistant soil and sustained food production. Food Agric. Org80.

  • Google Scholar

• 7

  BrentrupF.KüstersJ.KuhlmannH.LammelJ. (2004). Environmental impact assessment of agricultural production systems using the life cycle assessment methodology. Eur. J. Agron.20 (3), 247–264. 10.1016/s1161-0301(03)00024-8

  • CrossRef
  • Google Scholar

• 8

  CampbellB. M.ThorntonP.ZougmoréR.van AstenP.LipperL. (2017). Climate change adaptation and agriculture in sub-Saharan Africa: current and future research priorities. World Dev.98, 179–188.

  • Google Scholar

• 9

  ChaudhuriG.MishraN. B. (2016). Spatio-temporal dynamics of land cover and land surface temperature in Ganges-Brahmaputra delta: a comparative analysis between India and Bangladesh. Appl. Geogr.68, 68–83. 10.1016/j.apgeog.2016.01.002

  • CrossRef
  • Google Scholar

• 10

  DabalenA.GoyalA.SongR. (2024). Regenerative agriculture in practice: a review policy research working paper no. 10919.

  • Google Scholar

• 11

  European Commission (2015). Closing the loop - an EU action plan for the circular economy.

  • Google Scholar

• 12

  European Commission (2019). The European green deal.

  • Google Scholar

• 13

  GaribaldiL. A.Steffan-DewenterI.WinfreeR.AizenM. A.BommarcoR.CunninghamS. A.et al (2013). Wild pollinators enhance fruit set of crops regardless of honeybee abundance. Science339 (6127), 1608–1611. 10.1126/science.1230200

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  GavlakR.HorneckD.MillerR. O. (2005). Soil, plant and water reference methods for the western region. Western Region Extension Publication (WREP-125) (3rd ed.) Oregon State University.

  • Google Scholar

• 15

  GillerK. E.HijbeekR.AnderssonJ. A.SumbergJ. (2021). Regenerative agriculture: an agronomic perspective. Outlook Agric.50 (1), 13–25. 10.1177/0030727021998063

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  GodfrayH. C. J.BeddingtonJ. R.CruteI. R.HaddadL.LawrenceD.MuirJ. F.et al (2010). Food security: the challenge of feeding 9 billion people. Science327 (5967), 812–818. 10.1126/science.1185383

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  HillierJ.WalterC.MalinD.Garcia-SuarezT.Mila-i-CanalsL.SmithP. (2011). A farm-focused calculator for emissions from crop and livestock production. Environ. Model. Softw.26 (9), 1070–1078. 10.1016/j.envsoft.2011.03.014

  • CrossRef
  • Google Scholar

• 18

  JacksonM. L. (1973). Soil chemical analysis. Prentice Hall of India Pvt. Ltd.

  • Google Scholar

• 19

  JordonM. W.SmithP.LongP. R.BürknerP. C.PetrokofskyG.WillisK. J. (2022). Can regenerative agriculture increase national soil carbon stocks? Simulated country-scale adoption of reduced tillage, cover cropping, and ley-arable integration using RothC. Sci. Total Environ.825, 153955. 10.1016/j.scitotenv.2022.153955

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  JoseS. (2009). Agroforestry for ecosystem services and environmental benefits: an overview. Agrofor. Syst.76 (1), 1–10. 10.1007/978-90-481-3323-9_1

  • CrossRef
  • Google Scholar

• 21

  KabatoW.GetnetG. T.SinoreT.NemethA.MolnárZ. (2025). Towards climate-smart agriculture: strategies for sustainable agricultural production, food security, and greenhouse gas reduction. Agronomy15 (3), 565. 10.3390/agronomy15030565

  • CrossRef
  • Google Scholar

• 22

  KhanguraR.FerrisD.WaggC.BowyerJ. (2023). Regenerative agriculture—A literature review on the practices and mechanisms used to improve soil health. Sustainability15 (3), 2338. 10.3390/su15032338

  • CrossRef
  • Google Scholar

• 23

  KogaN.ShimodaS.ShiratoY.KusabaT.ShimaT.NiimiH.et al (2020a). Assessing changes in soil carbon stocks after land use conversion from forest land to agricultural land in Japan. Geoderma377, 114487. 10.1016/j.geoderma.2020.114487

  • CrossRef
  • Google Scholar

• 24

  KogaN.TashiroT.SawamotoT. (2020b). Soil carbon dynamics and mitigation strategies under different land use systems. Agric. Ecosyst. Environ.290, 106768.

  • Google Scholar

• 25

  KogloY. E.LalR.ShresthaR. K. (2016). Potential of soil carbon sequestration for climate change mitigation and sustainable agriculture. Environ. Syst. Res.5 (1), 1–14.

  • Google Scholar

• 26

  LalR. (2010). Enhancing eco-efficiency in agro-ecosystems through soil carbon sequestration. Crop Sci.50 (S1), S120–S131. 10.2135/cropsci2010.01.0012

  • CrossRef
  • Google Scholar

• 27

  MalhiS. S.LemkeR.WangZ. H.ChhabraB. S. (2006). Tillage, nitrogen and crop residue effects on crop yield, nutrient uptake, soil quality, and greenhouse gas emissions. Soil Tillage Res.90 (1-2), 171–183. 10.1016/j.still.2005.09.001

  • CrossRef
  • Google Scholar

• 28

  MarcacciG.KaboréR. R.ZongoA. N.ZoubgaS. T.DiakitéB.SpaarR.et al (2025). The potential of farmer-managed small-scale grazing exclusions as an adaptive management to restore degraded lands in arid regions. Biol. Conserv.302, 110949. 10.1016/j.biocon.2024.110949

  • CrossRef
  • Google Scholar

• 29

  MarcacciM.RossiR.BonfanteA. (2025). Soil pH and plant nutrient uptake: bridging knowledge for sustainable cropping systems. Eur. J. Agron.139, 126634.

  • Google Scholar

• 30

  MijatovićD.Van OudenhovenF.EyzaguirreP.HodgkinT. (2013). The role of agricultural biodiversity in strengthening resilience to climate change: towards an analytical framework. Int. J. Agric. Sustain.11 (2), 95–107. 10.1080/14735903.2012.691221

  • CrossRef
  • Google Scholar

• 31

  MishraA. K.DashP. K.MishraA.SharmaS. (2024). “Regenerative agriculture: a multifaceted approach to one health and soil restoration,” in Key drivers and indicators of soil health management: transitioning from conventional to regenerative agriculture (Singapore: Springer Nature Singapore), 1–32.

  • Google Scholar

• 32

  NelsonD. W.SommersL. E. (1996). Total carbon, organic carbon, and organic matter. In “Methods of soil analysis. Part 3: Chemical methods”, D. L. Sparks (Ed.), (Soil Science Society of America and American Society of Agronomy). 961–1010. 10.2136/sssabookser5.3.c34

  • CrossRef
  • Google Scholar

• 33

  NewtonP.CivitaN.Frankel-GoldwaterL.BartelK.JohnsC. (2020). What is regenerative agriculture? A review of scholar and practitioner definitions based on processes and outcomes. Front. Sustain. Food Syst.4, 577723. 10.3389/fsufs.2020.577723

  • CrossRef
  • Google Scholar

• 34

  OlsenS. R.ColeC. V.WatanabeF. S.DeanL. A. (1954). Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA Circular No. 939. U.S. Government Printing Office.

  • Google Scholar

• 35

  PadhiaryM.KumarR. (2024). “Assessing the environmental impacts of agriculture, industrial operations, and mining on agro-ecosystems,” in Smart internet of things for environment and healthcare (Cham: Springer Nature Switzerland), 107–126.

  • Google Scholar

• 36

  PandeyB.ZhangQ.SetoK. C. (2018). Time series analysis of satellite data to characterize multiple land use transitions: a case study of urban growth and agricultural land loss in India. J. Land Use Sci.13 (3), 221–237. 10.1080/1747423x.2018.1533042

  • CrossRef
  • Google Scholar

• 37

  PassiouraJ. B. (2006). Increasing crop productivity when water is scarce–from breeding to field management. Agric. Water Manag.80 (1-3), 176–196. 10.1016/j.agwat.2005.07.012

  • CrossRef
  • Google Scholar

• 38

  PathakH.ReddyY. R. (2021a). Soil carbon sequestration in Indian agriculture: challenges and opportunities. Curr. Sci.120 (6), 976–983.

  • Google Scholar

• 39

  PathakP.ReddyA. S. (2021b). Vertical distribution analysis of soil organic carbon and total nitrogen in different land use patterns of an agro-organic farm. Trop. Ecol.62, 386–397. 10.1007/s42965-021-00164-3

  • CrossRef
  • Google Scholar

• 40

  PooreJ.NemecekT. (2018). Reducing food’s environmental impacts through producers and consumers. Science360 (6392), 987–992. 10.1126/science.aaq0216

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  PotapovP.HansenM. C.PickensA.Hernandez-SernaA.TyukavinaA.TurubanovaS.et al (2022). The global 2000-2020 land cover and land use change dataset derived from the landsat archive: first results. Front. Remote Sens.3, 856903. 10.3389/frsen.2022.856903

  • CrossRef
  • Google Scholar

• 42

  RichardS.JamesD.StacieW. (2020). “InVEST 3.10.2,” User’s guide; the natural capital project; stanford university; university of Minnesota; the nature conservancy and world wildlife fund (Stanford, CA, USA).

  • Google Scholar

• 43

  SrokaW.DudekM.WojewodzicT.KrólK. (2019). Generational changes in agriculture: the influence of farm characteristics and socio-economic factors. Agriculture9 (12), 264. 10.3390/agriculture9120264

  • CrossRef
  • Google Scholar

• 44

  SubbiahB. V.AsijaG. L. (1956). A rapid procedure for the estimation of available nitrogen in soils. Current Science, 25, 259–260.

  • Google Scholar

• 45

  TothS. J.PrinceA. L. (1949). Estimation of cation-exchange capacity and exchangeable Ca, K, and Na contents of soils by flame photometer techniques. Soil Science, 67 (6), 439–445.

  • Google Scholar

• 46

  United Nations (2015). Paris agreement.

  • Google Scholar

• 47

  WillettW.RockströmJ.LokenB.SpringmannM.LangT.VermeulenS.et al (2019). Food in the anthropocene: the EAT–lancet commission on healthy diets from sustainable food systems. Lancet393 (10170), 447–492. 10.1016/s0140-6736(18)31788-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  YangX.XiongJ.DuT.JuX.GanY.LiS.et al (2024). Diversifying crop rotation increases food production, reduces net greenhouse gas emissions and improves soil health. Nat. Commun.15 (1), 198. 10.1038/s41467-023-44464-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  ZouY.ChenW.LiS.WangT.YuL.XuM.et al (2022). Spatio-temporal changes in vegetation in the last two decades (2001–2020) in the beijing–tianjin–hebei region. Remote Sens.14 (16), 3958. 10.3390/rs14163958

  • CrossRef
  • Google Scholar